| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
| |
This patch adds initial DCCP support for libnetfilter_conntrack.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
|
|
|
|
|
| |
This patch adds the size of the arrays to set to NULL unset
elements. This helps to spot unset functions for new attributes.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
|
|
|
|
|
|
| |
This patch adds support for explicit helper assignation. This support
will not be of any help without the appropriate kernel support that will
go into the Linux kernel 2.6.29 -sic-.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
|
|
|
|
|
|
| |
This patch cleanups the internal headers by splitting them into several
logical pieces.
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
- recover the ID support
- add support for timeout comparison
- ignore set operation for counters and use attributes
- fix broken status comparison
- statify several __snprintf functions
|
|
|
|
|
|
|
|
|
| |
so inserted conntracks get proper type, code and id. Without this fix
"type", "code" and "id" in reply direction were always set to 0. It
"automagically" worked for ICMP and ICMP_ECHOREPLY (ICMP_ECHOREPLY==8 ->
ICMP_ECHO==*0*), but not with with other ICMP codes nor with ICMPv6.
Signed-off-by: Krzysztof Oledzki <ole@ans.pl>
|
| |
|
| |
|
| |
|
| |
|
|
|
|
|
|
|
|
| |
ICMP ID is stored as a u_int16_t, but its setter function derefs it's
arguement as a u_int8_t. Additionally the api "doc" claims it's a u8, when
it's not.
This patch fixes both.
|
|
|
|
|
|
| |
status flags bits of conntrack objects in userspace can be set and unset as it happens with other attributes.
- nfct_get_objopt with NAT detectors previously checks if the status attribute is set, otherwise it just skips it.
|
|
- object oriented infrastructure
- extensible and configurable output (XML)
- low level functions to interact with netlink details
- fairly documented
Still backward compatible.
|